Olympic figure skating looks effortless. Athletes swim across the ice, then fly into the air, spinning like a top, before landing on a blade only 4–5 millimeters wide. To help figure skaters land the quadruple Axel, Salchow, Lutz and perhaps the elusive quintuple without the slightest strain, Jerry Lou Mfin ’24 developed an optical tracking system called OOFSkate Which uses artificial intelligence to analyze videos of figure skaters’ jumps and make recommendations on how to improve. Lu, a former researcher MIT Sports Labare assisting Team USA’s elite skaters with their technical performances and will work with NBC Sports during the 2026 Winter Olympics to help commentators and TV viewers better understand the complex scoring systems in figure skating, snowboarding and skiing. He will use AI technologies to explain the nuances of judging decisions and demonstrate how technically challenging these games can be.
Meanwhile, Professor Annette “Peko” Hosoi, co-founder and faculty director of the MIT Sports Lab, is working on new research aimed at understanding how AI systems evaluate aesthetic performances in figure skating. Hosoi and Lu recently spoke MIT News About implementing AI in sports, whether AI systems could be used to evaluate Olympic figure skating, and when we might see a skater descend to a quint.
Why: Why Apply AI to Figure Skating?
Loo: Skaters can always push, higher, faster, stronger. OOFSkate aims to help skaters find a way to spin a little faster or jump a little higher in their jumps. This system helps skaters catch things that might pass the eye test, but also allows them to target some higher-value areas of opportunity. The artistic side of skating is more difficult to evaluate than the technical elements because it is subjective.
To use the mobile training app, all you need to do is take a video of an athlete jumping, and it will report physical metrics that tell you how many revolutions you can perform. It tracks those metrics and builds in others for all current elite and former elite athletes. You can look at your data and then see, “This is how an Olympic champion did this element, maybe I should try that.” You get comparisons and an automated classifier that shows you if you did this trick at the World Championships and it was judged by an international panel, this is approximately the grade of execution score they would give you.
Hosoi: There are a lot of AI tools coming online, especially things like pose estimators, where you can infer skeletal configuration from video. The challenge with these pose estimators is that if you only have one camera angle, they perform very well at camera level, but with depth they perform very poorly. For example, if you’re trying to critique someone’s form in fencing, and they’re moving towards the camera, you get very poor data. But with figure skating, Jerry has found one of the few fields where depth challenges don’t really matter. In figure skating, you need to understand: how high did this person jump, how many times did they go around, and how well did they land? None of these are depth dependent. They found an application that makes estimators perform really well, and they don’t have to pay any penalty for the work they do poorly.
Why: Could you ever see a world in which AI is used to evaluate the artistic side of figure skating?
Hosoi: When it comes to AI and beauty assessment, we have new work underway thanks to an MIT Human Insight Collaborative (MITHIC) grant. This work is a collaboration with Professor Arthur Bahr and IDSS graduate student Eric Liu. When you ask an AI platform for an aesthetic assessment like “What do you think of this painting?” It will react with something that sounds like it came from a human. We want to understand that, to reach that evaluation, are AIs going through the same reasoning pathways or using the same intuitive concepts that humans go through to reach the conclusion, “I like that painting,” or “I don’t like that painting”? Or are they just parrots? Are they just copying what they heard someone say? Or is there a concept map of aesthetic appeal? Figure skating is an ideal place to look at this map because skating is judged aesthetically. And there are numbers. You can’t go to a museum and say, “This painting is 35.” But in skating, you have data.
This brings up another interesting question, which is the difference between beginners and experts. It is known that expert humans and novice humans will react differently after seeing the same thing. Someone who is an expert judge may have a different opinion about skating performance than a member of the general population. We’re trying to understand the differences between the responses of experts, novices, and AI. Is there any similarity in these reactions in terms of where they are coming from, or is the AI coming from a different place than both the expert and the novice?
Loo: Figure skating is interesting because everyone working in the field of AI is trying to figure out AGI or artificial general intelligence and trying to build this extremely strong AI that mimics humans. Working on applying AI to sports like figure skating helps us understand how humans think and make decisions. This has a direct impact on AI research and companies that develop AI models. Gaining a deep understanding of how current state-of-the-art AI models work with these games, and how you need to train and fine-tune these models to work for specific games, helps you understand how AI needs to move forward.
Why: Now that you are studying and working in this field, what will you be seeing at the Milan Cortina Olympic figure skating competitions? Do you think anyone will even get a quint?
Loo: For winter sports, I’m working with NBC for figure skating, ski and snowboarding competitions to help them tell a data-driven story for the American people. The goal is to make these games more relevant. Skating looks slow on television, but it is not. It is believed that everything will look smooth. If it seems difficult, you will probably be penalized. Skaters need to learn how to spin very fast, jump very high, float in the air and land beautifully on one foot. The data we’re collecting can help show how hard skating really is, even when it seems easy.
I’m glad we’re working in the Olympic sports arena because the world watches the Olympics once every four years, and it’s traditionally a coaching-intensive and talent-driven sport, unlike a sport like baseball, where if you don’t have an elite level optical tracking system you’re not maximizing the value of what you currently have. I’m glad we got the chance to work with these Olympic Games and athletes and make an impact here.
Hosoi: Ever since I had a chance to turn on the TV, I have always watched the Olympic figure skating competitions. They are always incredible. One of the things I’m going to practice is identifying jumps, which is very hard to do if you’re an amateur “judge.”
I even did some back of the envelope calculations to see if a quint was possible. I am now completely convinced that it is possible. We will see it within our lifetime, if not relatively soon. Not in this Olympics, but soon. When I saw we were so close on quint, I thought, what about six? Can we do six rounds? Probably not. This is where we start to come up against the limits of human physical capacity. But I think five is within reach.
